[Insight-developers] SimpleITK Community Survey Draft
Caban, Jesus (NIH/NLM/LHC) [F]
jesus.caban at nih.gov
Fri Sep 17 08:54:15 EDT 2010
Hi Gabe,
The survey looks great! Thanks for making all the changes!
I have one minor suggestion / comment. The only question that I found that might cause some confusion is
Q8: "How important is a short compilation time for SimpleITK?". There are two ways to look at this: (a) initial
compilation time (for those using ITK from source code) or (b) compilation time when using SimpleITK within
an application. I think we are referring to (a), right? If so, maybe we can modify the question to specify
that it's the first time around.
In general, I think the survey is good to go. Once you remove all the "test responses", please let us know
so we can start sharing the survey...
Thanks!
-J
On Sep 16, 2010, at 3:52 PM, Gabe Hart wrote:
> Hi Jesus,
>
> More great comments. I just went through them and made a bunch of
> changes. As far as the possibility of adding logic to the survey,
> SurveyMonkey does support very primitive logic, but it seems to be based
> around the concept of one question per page (as in you can change what
> the next page is based on a response on the current page). I personally
> think that this paradigm of one question per page makes the survey feel
> much longer and should be avoided. As such, I've tried to collect most
> of the questions that are only valid for people who have used ITK before
> on the last page so that the entire page can be skipped if desired.
>
> Also, none of the questions are currently mandatory, so any that don't
> apply can be skipped (maybe I should add a note about this in the
> intro). Does this make sense to everyone? Should we try to find a
> subset that should be required, or is okay to just get any results the
> users want to offer?
>
> As for SurveyMonkey, we're currently using the Pro version which allows
> up to 1000 responses per month with a $.10 overage for each additional
> response. The next option up allows unlimited responses but requires a
> minimum of a year's subscription.
>
> -Gabe
>
> On 09/16/2010 12:36 PM, Caban, Jesus (NIH/NLM/LHC) [F] wrote:
>> Hi everyone,
>> This is great! Overall, I think this is a great start. However, I don't know if the survey is ready for prime-time yet. Since with SimpleITK we are trying in expand the ITK user base, I think we should spend more time polishing the survey. That includes distributing the survey among a handful of non-ITK developers and non-computer scientist (e.g. radiologists) and get their feedback about the survey (before opening it to the public).
>>
>> Here are some of my comments/suggestions:
>>
>> Section 1:
>> Q4: "What programming tools do you use?"
>> Maybe we should be more specific by saying something like "What programming tools do you use to process your images?"
>>
>> Q5: "Which of the following best describes why you write code?"
>> Maybe it's related to option #3 "a scientific application", but maybe we can add something like "I write code to prototype new algorithms, techniques and/or applications".
>>
>>
>>
>> Section 3: SimpleITK Questions
>> Intro: "These questions focus on SimpleITK. The goal is to figure... ". I think this intro is too informal. Since we are distributing this survey to other groups (radiologists, scientists, etc...) we need to make it look more professional. Maybe something like "The purpose of the following questions is to gather information about how you plan to use ITK and what feature you are more likely to use..."
>>
>>
>> Q1: "larger than RAM?"
>> Maybe larger than local memory instead of RAM?
>>
>>
>> Q2. "How many dimensions do you need image to have?"
>> This is not clear. Maybe: "What image dimensionalities(maybe another word) are you more likely to use? Also, instead of just 1,2,3..., we can say 1D, 2D, 3D, 4D, other...
>>
>>
>> Q5: We should add "others" to this list. Some researchers might use cimg, PIL, etc...
>>
>>
>> Q5: All the examples are assuming that people are familiar with C++. (e.g. itk::Image) I think nonITK programmers might understand better the pseudo code that we have on the wiki page. For example, to understand the following code you don't need to know c++
>>
>> ITKImage im = ImageFileReader ("sample/path/to/image.jpg");
>> im = GaussianFilter (im, 2);
>> ImageFileWriter (im, "sample/path/to/output.png");
>>
>>
>> Q5: We should add something like "Don't understand the code" or "Don't understand the differences" under the possible selections
>>
>> Q2. "How helpful do you find the current ITK error"
>> What about if they haven't used ITK? We should omit this question for people that said they have never used ITK.
>>
>>
>>
>> OTHERS:
>> 1) Should we add a question about how important is DICOM/PACs support within SimpleITK? (i.e. retrieve, push, etc..)
>>
>> 2) A number of the questions can be omitted if they haven't used ITK. Can we do branches (if statements) with SurveyMonkey?
>>
>> 3) At the beginning of the survey, should we specify that this is an anonymous survey?
>>
>> 4) Are we using the "free" version of SurveyMonkey, what's the maximum number of responses we can get? Any limits?
>>
>>
>> Thanks!
>> -Jesus Caban
>>
>>
>>
>>
>> On Sep 16, 2010, at 11:03 AM, Gabe Hart wrote:
>>
>>
>>> After a couple of good sets of feedback, I think the survey is ready to
>>> go. The link again is:
>>>
>>> http://www.surveymonkey.com/s/SimpleITKCommunitySurvey
>>>
>>> If anyone has any more comments, feel free to send them to me. My goal
>>> is to have a fully finished version by the end of the day today so that
>>> we can start distributing it tomorrow.
>>>
>>> -Gabe
>>>
>>> On 09/16/2010 01:50 AM, Dan Mueller wrote:
>>>
>>>> Hi Gabe,
>>>>
>>>> Thanks for creating the survey. Some thoughts:
>>>>
>>>> Page 1, "ITK is being significantly revamped for the upcoming version 4 release"
>>>> Consider adding the expected release date in brackets.
>>>>
>>>> Page 1, "One of the most important pieces of feedback from our users
>>>> has been that ITK's power comes with a steep learning curve and makes
>>>> many less-complicated tasks more difficult than necessary."
>>>> This sentence is a bit wordy. People are busy, consider shortening it.
>>>> For example:
>>>> "Previous user feedback has indicated ITK's power comes with a steep
>>>> learning curve, sometimes over-complicating simple tasks."
>>>>
>>>> Page 1, "The goal of this survey is to collect your thoughts on what
>>>> the most important features of ITK's new simple layer should be."
>>>> Again, perhaps a bit wordy. Consider:
>>>> "The aim of this survey is to determine the most important features to
>>>> include in the new ITK simple layer."
>>>>
>>>> Page 2, question 5 "Which of the following best describes your why you
>>>> write code?"
>>>> Typo: remove "your".
>>>>
>>>> Page 2, question 8. What operating systems do you use for image analysis work?
>>>> Do we need an "other" free-text field?
>>>>
>>>> Page 2, questions 10,11,12: First time I read these questions, I
>>>> thought they were the same on (a typo). Is it possible to underline
>>>> "understandable", "easy", "useful" to clearly show the difference? Do
>>>> we really need to ask all three?
>>>>
>>>> Page 3, question 9. What types of devices would you like to be able to
>>>> use SimpleITK on?
>>>> Do we need an "other" free-text field? Mobile device?
>>>>
>>>> Page 4, question 3. How important is it to you for SimpleITK to use a
>>>> pipeline system similar to the one in ITK?
>>>> Can this question be included in question 2 above?
>>>> For example:
>>>> Simple image filters (eg. Gaussian filter)
>>>> Pipeline of image filters (eg. Gaussian then Sigmoid filter)
>>>>
>>>> HTH
>>>>
>>>> Cheers, Dan
>>>>
>>>> On 15 September 2010 23:14, Gabe Hart<gabe.hart at kitware.com> wrote:
>>>>
>>>>
>>>>> Hi all,
>>>>>
>>>>> Here is the link to the first draft of the SimpleITK community survey:
>>>>>
>>>>> http://www.surveymonkey.com/s/SimpleITKCommunitySurvey
>>>>>
>>>>> If anyone has feedback on changes (new questions, rewording, order, etc...),
>>>>> please let me know and I'll make the appropriate adjustments. Also, an
>>>>> initial list of target communities has been compiled at
>>>>>
>>>>> http://www.itk.org/Wiki/ITK_Release_4/SimpleITK/Survey_1#Target_Audience
>>>>>
>>>>> If anyone knows of other communities that should be added, please put them
>>>>> up on the wiki. Our goal is to find someone who is already involved in each
>>>>> of these groups to publish the surveys, so if you or someone you know would
>>>>> be a good liaison, please put that on the wiki as well.
>>>>>
>>>>> Thanks,
>>>>> -Gabe
>>>>>
>>>>>
>>>>>
>>>
>>
>
More information about the Insight-developers
mailing list